| Abstract
| - Basing clinical practice decisions on currently available evidence from clinical trials may result in suboptimal care in the real-world practice of medicine. The problem stems from both the quality of data and the means by which the data have been interpreted. In an attempt to remedy the situation, a term based on an assessment of all risk factors and comorbidity has been suggested and made more easily translatable to clinical decision making. This term, the reciprocal of absolute risk reduction, is the number needed to treat (NNT), an idea that has gained wide acceptance. There are, however, risks that the simplicity of the NNT will be misused to guide reimbursement decisions and potentially prevent patients from receiving optimal care. This seemingly objective measure may be seriously flawed if the data on which this measure is based do not necessarily reflect the results that real world doctors achieve in practice. These problems could lead to the misguided allocation of health care resources. It is therefore incumbent upon us to closely evaluate the data on which NNTs are based and, if necessary, to arrive at more accurate NNTs. In our view, such data should be gleaned from effectiveness trials done by real world doctors in real world settings with real world volunteers. Large simple trials, as have already been performed in a number of therapeutic areas, especially for acute management of myocardial infarction and in acquired immune deficiency syndrome, offer the best likelihood of yielding this crucial information. We need to be sure that future trials of chronic conditions such as hypertension are done with this trial paradigm, so that those who pay for care have the accurate knowledge needed to spend their money wisely.
|
